Chapter 5 Results
5.1 New York City’s Arrests
In the result section, you will find visualizations, and the insights derived from the cleaned & preprocessed NYC Arrest dataset, organized around the concepts of 3 Ws: What, Where, and When.
5.1.1 What:
5.1.1.1 What is the distribution of Arrest Types?
First of all, let’s use a bar plot to visualize the general distribution of arrest category. Since there are a lot of arrest types, I would only show the top 15 counts of the arrest types. In this way, we are able to visualize what kind of crimes are happening and being arrested in NYC.
From the plot, we can observe that the highest number of arrests made by NYPD is to the type of “Dangerous Drugs”, having 466735 cases, significantly higher than the counts of rest of types. Indeed, NYC always has a bad reputation with dangerous drugs. The second highest count is the type of “Assault 3 & Related OFFENSES”, which suggests the relative safety problem faced by NYC residents every day. We need to have tighter laws and control towards drugs as this is the most serious and frequent arrest categories.
Next, let’s investigate the distribution of arrests count throughout NYC’s 5 boroughs.
5.1.1.2 What is the distribution of arrest rate in each NYC boroughs?
In the above bar plot, we calculated the average arrest rate of NYC residents across the ten year interval, from 2012-2021. The plot, suggests that for each year, residents in Manhattan will be arrested for an average of 0.046 times, the highest rate across the five boroughs. Bronx is the second highest with a rate of 0.045. Both rates are significantly higher than the rate of the rest of 3 boroughs.
Netx, let’s check the
5.1.1.3 What kind of crimes do different age groups tend to commit?


First of all, for residents whose age are older than 65, they barely commit any crimes or being arrested by NYPD. As expected, the arrests made by NYPD are heavily concentrated on the age group of 25-44. The second highest age group is 18-24. It indicates that most crimes committed and offenders arrested are in a relatively young age. For the distribution of offenders’ age group younger than 65 vs arrest categories, we find that for age group from 18-64, “dangerous drugs” has the highest number of arrests made, which is consistent with the overall arrest categories distribution. For teenager who is under 18, however, “robbery”, “assault” and “dangerous drugs” all have similar numbers of arrests, with robbery being the slightly highest. Unlike for other age groups, robbery stands out to be the most frequent one. This is alarming.
We need to build better programs and have better education to get more teenagers off the streets and back to school.
For the distribution of offenders’ race vs arrest categories, first we found that there are barely any arrests made to “American Indian / Alaskan Native”, presumably because of the low populations of the group living in New York City. For the rest of the races, we can observe that, except for “Asian / Pacific Islander”, all the offenders in the other races were being arrested due to “dangerous drugs” most frequently. For “Asian / Pacific Islander” residents, the highest arrest category is “Assault 3 & Related Offenses”. Additionally, there is an alarming trend: The number of arrests made to black NYC residents is significantly higher than that of any other races. Is there a systematic discrimination against black people in the arrests made by NYPD? In the When section, we will use time series data to find out. However, better education and benefits program are necessary to solve this situation.
For the distribution of offenders’ sex vs arrest categories, intuitively the number of male offenders are significantly higher than the number of female offenders. Interetsingily, for female offenders, the highest arrests made are to the category of “Assault 3 & Related Offenses”, with “Petit Larceny” and “Dangerous Drugs” being closer for the second and third. On the other hand, male offenders in NYC are usually arrested for “Dangerous Drugs”, which is consistent with the overall distribution.
In the next section, let’s use some maps to explore and visualize the locations as well as each arrests being made.
5.1.2 Where:
We will use New Yorkc City’s map to visualize the type and location of the arrests occurred. For normal citizens, they are more concerned about crimes of theft, assault, weapons as well as robbery. Thus, we will create a map to visualize the frequencies of occurrences of these types of arrests.